Explain Decision (Grade A)Habilidade de dados-ai testada de segurança para Claude AI. Grau A. Recuperar o raciocínio completo por trás de uma decisão de política do AxonFlow — políticas correspondentes, nível de risco, disponibilidade de substituição, contagem de acertos recentes

What gets checked

AxonFlow ships with 80+ built-in system policies that apply to Claude Code automatically. No configuration required — new policies added to the platform are immediately enforced in every session. | Category | Coverage | |---|---| | Dangerous commands | Reverse shells (nc -e, bash -i, /dev/tcp/), rm -rf /, dd if=, curl \| bash, credential file access (cat ~/.ssh/, cat ~/.aws/), path traversal | | SQL injection | 30+ patterns including UNION injection, stacked queries, auth bypass, encoding tricks | | PII detection | SSN, credit card, Aadhaar, PAN, email, phone, NRIC/FIN (Singapore), and more — with redaction | | Secrets exposure | API keys, connection strings, hardcoded credentials, code secrets | | SSRF | Cloud metadata endpoint (169.254.169.254) and internal-network blocking | | Prompt injection | Instruction override, jailbreak attempts, role hijacking | | Claude Code-specific | .claude/settings.json write protection, .claude/hooks/.json modification warnings (enabled via AXONFLOWINTEGRATIONS=claude-code) | Custom policies are easy — POST /api/v1/dynamic-policies or the Customer Portal. See Policy Enforcement. ---

Asked by Lena Fischer · Mar 31, 2026

How it works

Governed tools: Bash, Write, Edit, NotebookEdit, and all MCP server tools (mcp). Read-only tools (Read, Glob, Grep) are not governed by default — they don't modify state or send data externally. Fail behavior:** AxonFlow unreachable (network) → fail-open, tool execution continues AxonFlow auth/config error → fail-closed, tool call denied until config is fixed PostToolUse failures → never block (audit and PII scan are best-effort) ---

Asked by Adaeze Uche · Mar 22, 2026

Why you'd add this

Claude Code is Anthropic's official CLI — a fast, agentic coding assistant that edits files, runs shell commands, and calls MCP servers. It's excellent at developer productivity. It was never designed to be the layer where your security and compliance team lives. The gaps start surfacing the moment Claude Code moves from one developer's laptop to a team or production setting: | Production requirement | Claude Code alone | With this plugin | |---|---|---| | Policy enforcement before tool execution | Hooks available, no governance logic | 80+ built-in policies evaluated on every governed tool call | | Dangerous command blocking (rm -rf /, reverse shells, curl \| bash) | Not addressed | Blocked before execution with decision context | | PII / secrets detection in tool outputs | Developer responsibility | Auto-scan; Claude is instructed to use redacted version | | SQL-injection detection on MCP queries | MCP server's problem | 30+ patterns evaluated on every MCP tool call | | Compliance-grade audit trail | Session logs, not compliance-formatted | Every governed call recorded with policies, decision, duration | | Decision explainability after a block | Generic hook failure message | decisionid surfaced in deny reason; explaindecision MCP tool returns the full record | | Self-service, time-bounded exceptions | Not available | createoverride with mandatory justification, fully audited | | Cloud metadata / SSRF / path traversal blocking | Not addressed | Built in** | You get all of t

Asked by Hana Kobayashi · Jan 29, 2026